Last week, I spoke with Chris Farrell of MPR News...
%20(1).webp)
This blog post is based on a conversation CEO Senan Ebrahim, MD, PhD had on MPR news with Chris Farrell. Listen to their entire conversation here.
Last week, I spoke with Chris Farrell of MPR News and Dr. Christopher Tignanelli about AI in healthcare. We are in the midst of a healthcare revolution: this era of computing has democratized access to AI technology, allowing people all over the world to train and run algorithms to improve care. Here are some highlights of the insights Dr. Tignanelli and I shared with Chris. For our full conversation, listen here.
I believe that AI has the potential to massively augment the abilities of clinicians to deliver better care faster. Beyond the quintessential examples of machine learning-based binary classifiers, an unsupervised algorithm can look at data at a scale that humans cannot typically process, and identify new insights. For example, during the initial stages of the COVID-19 pandemic, researchers applied AI to clinical notes to identify anosmia—loss of smell—as a specific symptom that suggested infection.
AI has the potential to truly personalize a patient’s care. We have not even scratched the surface of how AI can leverage vast amounts of data to create a clear and vibrant data picture for every patient. By combining insights from health data including clinical notes, genomic data, imaging data, and physiologic data, we can enable clinicians to make better decisions about each patient’s care more accurately and efficiently.
Multiple callers expressed well-founded concerns about the use of AI in healthcare. Some fear that AI will destroy their personal connection with their provider. I believe that connection is sacrosanct. That therapeutic doctor-patient relationship is the reason that so many of us got into healthcare in the first place. As I see it, the role of AI is to augment that relationship—not detract from it. Another caller shared the concerning fact that general AI models like ChatGPT also “hallucinate” to invent nonexistent sources and information. No physicians or scientists I know are using GPT models in any kind of informational capacity. That said, there are already companies effectively addressing the hallucination problem, and in the next five years we will likely have models that reliably deliver factually accurate information.
Another risk of AI in healthcare is encoded bias. In pregnancy care, we think about bias a lot, as the brunt of the maternal health crisis has fallen largely on the shoulders of Black and Native women, who have around 3x worse outcomes than white women in the U.S. At Delfina, when we build our models to serve these populations, we mitigate bias of the data inputs—including selection bias, as in whose data we are including, and any racial bias in how that data is collected or presented.
With AI, we have a compelling opportunity to flip the script from reactive medicine to preventative care. Especially in prenatal care, OBs are under financial pressure to see more and more patients a day, spending less time with each. Though well-intentioned, previous attempts at value-based care have fallen flat due to maternal care deserts, and physicians lacking the tools to efficiently offer each patient the whole-person care that they need. With the kinds of AI-driven tools we’re developing at Delfina, we believe that we can help reorient the healthcare system to truly adopt an outcomes-driven approach, one without the financial imperatives to grow volume or squeeze margins.
Though the use of AI in maternal health, and in healthcare in general, is still in the early stages, I believe that these tools will be monumental in transforming healthcare for the better. I am excited to see physicians, academics, entrepreneurs, and the government come together to leverage these new technologies for the benefit of patients.
I am grateful to Dr. Tignanelli, Chris Farrell, and all our callers on the program for engaging in this discussion with me. There is so much more to this timely topic, so if you are curious about AI in healthcare, listen to our full conversation here. If you want to learn more about how we are leveraging AI technology at Delfina to deliver customized, closed-loop care to pregnant patients, join us.
Last week, I spoke with Chris Farrell of MPR News...
This blog post is based on a conversation CEO Senan Ebrahim, MD, PhD had on MPR news with Chris Farrell. Listen to their entire conversation here.
Last week, I spoke with Chris Farrell of MPR News and Dr. Christopher Tignanelli about AI in healthcare. We are in the midst of a healthcare revolution: this era of computing has democratized access to AI technology, allowing people all over the world to train and run algorithms to improve care. Here are some highlights of the insights Dr. Tignanelli and I shared with Chris. For our full conversation, listen here.
I believe that AI has the potential to massively augment the abilities of clinicians to deliver better care faster. Beyond the quintessential examples of machine learning-based binary classifiers, an unsupervised algorithm can look at data at a scale that humans cannot typically process, and identify new insights. For example, during the initial stages of the COVID-19 pandemic, researchers applied AI to clinical notes to identify anosmia—loss of smell—as a specific symptom that suggested infection.
AI has the potential to truly personalize a patient’s care. We have not even scratched the surface of how AI can leverage vast amounts of data to create a clear and vibrant data picture for every patient. By combining insights from health data including clinical notes, genomic data, imaging data, and physiologic data, we can enable clinicians to make better decisions about each patient’s care more accurately and efficiently.
Multiple callers expressed well-founded concerns about the use of AI in healthcare. Some fear that AI will destroy their personal connection with their provider. I believe that connection is sacrosanct. That therapeutic doctor-patient relationship is the reason that so many of us got into healthcare in the first place. As I see it, the role of AI is to augment that relationship—not detract from it. Another caller shared the concerning fact that general AI models like ChatGPT also “hallucinate” to invent nonexistent sources and information. No physicians or scientists I know are using GPT models in any kind of informational capacity. That said, there are already companies effectively addressing the hallucination problem, and in the next five years we will likely have models that reliably deliver factually accurate information.
Another risk of AI in healthcare is encoded bias. In pregnancy care, we think about bias a lot, as the brunt of the maternal health crisis has fallen largely on the shoulders of Black and Native women, who have around 3x worse outcomes than white women in the U.S. At Delfina, when we build our models to serve these populations, we mitigate bias of the data inputs—including selection bias, as in whose data we are including, and any racial bias in how that data is collected or presented.
With AI, we have a compelling opportunity to flip the script from reactive medicine to preventative care. Especially in prenatal care, OBs are under financial pressure to see more and more patients a day, spending less time with each. Though well-intentioned, previous attempts at value-based care have fallen flat due to maternal care deserts, and physicians lacking the tools to efficiently offer each patient the whole-person care that they need. With the kinds of AI-driven tools we’re developing at Delfina, we believe that we can help reorient the healthcare system to truly adopt an outcomes-driven approach, one without the financial imperatives to grow volume or squeeze margins.
Though the use of AI in maternal health, and in healthcare in general, is still in the early stages, I believe that these tools will be monumental in transforming healthcare for the better. I am excited to see physicians, academics, entrepreneurs, and the government come together to leverage these new technologies for the benefit of patients.
I am grateful to Dr. Tignanelli, Chris Farrell, and all our callers on the program for engaging in this discussion with me. There is so much more to this timely topic, so if you are curious about AI in healthcare, listen to our full conversation here. If you want to learn more about how we are leveraging AI technology at Delfina to deliver customized, closed-loop care to pregnant patients, join us.
Last week, I spoke with Chris Farrell of MPR News...
This blog post is based on a conversation CEO Senan Ebrahim, MD, PhD had on MPR news with Chris Farrell. Listen to their entire conversation here.
Last week, I spoke with Chris Farrell of MPR News and Dr. Christopher Tignanelli about AI in healthcare. We are in the midst of a healthcare revolution: this era of computing has democratized access to AI technology, allowing people all over the world to train and run algorithms to improve care. Here are some highlights of the insights Dr. Tignanelli and I shared with Chris. For our full conversation, listen here.
I believe that AI has the potential to massively augment the abilities of clinicians to deliver better care faster. Beyond the quintessential examples of machine learning-based binary classifiers, an unsupervised algorithm can look at data at a scale that humans cannot typically process, and identify new insights. For example, during the initial stages of the COVID-19 pandemic, researchers applied AI to clinical notes to identify anosmia—loss of smell—as a specific symptom that suggested infection.
AI has the potential to truly personalize a patient’s care. We have not even scratched the surface of how AI can leverage vast amounts of data to create a clear and vibrant data picture for every patient. By combining insights from health data including clinical notes, genomic data, imaging data, and physiologic data, we can enable clinicians to make better decisions about each patient’s care more accurately and efficiently.
Multiple callers expressed well-founded concerns about the use of AI in healthcare. Some fear that AI will destroy their personal connection with their provider. I believe that connection is sacrosanct. That therapeutic doctor-patient relationship is the reason that so many of us got into healthcare in the first place. As I see it, the role of AI is to augment that relationship—not detract from it. Another caller shared the concerning fact that general AI models like ChatGPT also “hallucinate” to invent nonexistent sources and information. No physicians or scientists I know are using GPT models in any kind of informational capacity. That said, there are already companies effectively addressing the hallucination problem, and in the next five years we will likely have models that reliably deliver factually accurate information.
Another risk of AI in healthcare is encoded bias. In pregnancy care, we think about bias a lot, as the brunt of the maternal health crisis has fallen largely on the shoulders of Black and Native women, who have around 3x worse outcomes than white women in the U.S. At Delfina, when we build our models to serve these populations, we mitigate bias of the data inputs—including selection bias, as in whose data we are including, and any racial bias in how that data is collected or presented.
With AI, we have a compelling opportunity to flip the script from reactive medicine to preventative care. Especially in prenatal care, OBs are under financial pressure to see more and more patients a day, spending less time with each. Though well-intentioned, previous attempts at value-based care have fallen flat due to maternal care deserts, and physicians lacking the tools to efficiently offer each patient the whole-person care that they need. With the kinds of AI-driven tools we’re developing at Delfina, we believe that we can help reorient the healthcare system to truly adopt an outcomes-driven approach, one without the financial imperatives to grow volume or squeeze margins.
Though the use of AI in maternal health, and in healthcare in general, is still in the early stages, I believe that these tools will be monumental in transforming healthcare for the better. I am excited to see physicians, academics, entrepreneurs, and the government come together to leverage these new technologies for the benefit of patients.
I am grateful to Dr. Tignanelli, Chris Farrell, and all our callers on the program for engaging in this discussion with me. There is so much more to this timely topic, so if you are curious about AI in healthcare, listen to our full conversation here. If you want to learn more about how we are leveraging AI technology at Delfina to deliver customized, closed-loop care to pregnant patients, join us.
Last week, I spoke with Chris Farrell of MPR News...
This blog post is based on a conversation CEO Senan Ebrahim, MD, PhD had on MPR news with Chris Farrell. Listen to their entire conversation here.
Last week, I spoke with Chris Farrell of MPR News and Dr. Christopher Tignanelli about AI in healthcare. We are in the midst of a healthcare revolution: this era of computing has democratized access to AI technology, allowing people all over the world to train and run algorithms to improve care. Here are some highlights of the insights Dr. Tignanelli and I shared with Chris. For our full conversation, listen here.
I believe that AI has the potential to massively augment the abilities of clinicians to deliver better care faster. Beyond the quintessential examples of machine learning-based binary classifiers, an unsupervised algorithm can look at data at a scale that humans cannot typically process, and identify new insights. For example, during the initial stages of the COVID-19 pandemic, researchers applied AI to clinical notes to identify anosmia—loss of smell—as a specific symptom that suggested infection.
AI has the potential to truly personalize a patient’s care. We have not even scratched the surface of how AI can leverage vast amounts of data to create a clear and vibrant data picture for every patient. By combining insights from health data including clinical notes, genomic data, imaging data, and physiologic data, we can enable clinicians to make better decisions about each patient’s care more accurately and efficiently.
Multiple callers expressed well-founded concerns about the use of AI in healthcare. Some fear that AI will destroy their personal connection with their provider. I believe that connection is sacrosanct. That therapeutic doctor-patient relationship is the reason that so many of us got into healthcare in the first place. As I see it, the role of AI is to augment that relationship—not detract from it. Another caller shared the concerning fact that general AI models like ChatGPT also “hallucinate” to invent nonexistent sources and information. No physicians or scientists I know are using GPT models in any kind of informational capacity. That said, there are already companies effectively addressing the hallucination problem, and in the next five years we will likely have models that reliably deliver factually accurate information.
Another risk of AI in healthcare is encoded bias. In pregnancy care, we think about bias a lot, as the brunt of the maternal health crisis has fallen largely on the shoulders of Black and Native women, who have around 3x worse outcomes than white women in the U.S. At Delfina, when we build our models to serve these populations, we mitigate bias of the data inputs—including selection bias, as in whose data we are including, and any racial bias in how that data is collected or presented.
With AI, we have a compelling opportunity to flip the script from reactive medicine to preventative care. Especially in prenatal care, OBs are under financial pressure to see more and more patients a day, spending less time with each. Though well-intentioned, previous attempts at value-based care have fallen flat due to maternal care deserts, and physicians lacking the tools to efficiently offer each patient the whole-person care that they need. With the kinds of AI-driven tools we’re developing at Delfina, we believe that we can help reorient the healthcare system to truly adopt an outcomes-driven approach, one without the financial imperatives to grow volume or squeeze margins.
Though the use of AI in maternal health, and in healthcare in general, is still in the early stages, I believe that these tools will be monumental in transforming healthcare for the better. I am excited to see physicians, academics, entrepreneurs, and the government come together to leverage these new technologies for the benefit of patients.
I am grateful to Dr. Tignanelli, Chris Farrell, and all our callers on the program for engaging in this discussion with me. There is so much more to this timely topic, so if you are curious about AI in healthcare, listen to our full conversation here. If you want to learn more about how we are leveraging AI technology at Delfina to deliver customized, closed-loop care to pregnant patients, join us.
Last week, I spoke with Chris Farrell of MPR News...
This blog post is based on a conversation CEO Senan Ebrahim, MD, PhD had on MPR news with Chris Farrell. Listen to their entire conversation here.
Last week, I spoke with Chris Farrell of MPR News and Dr. Christopher Tignanelli about AI in healthcare. We are in the midst of a healthcare revolution: this era of computing has democratized access to AI technology, allowing people all over the world to train and run algorithms to improve care. Here are some highlights of the insights Dr. Tignanelli and I shared with Chris. For our full conversation, listen here.
I believe that AI has the potential to massively augment the abilities of clinicians to deliver better care faster. Beyond the quintessential examples of machine learning-based binary classifiers, an unsupervised algorithm can look at data at a scale that humans cannot typically process, and identify new insights. For example, during the initial stages of the COVID-19 pandemic, researchers applied AI to clinical notes to identify anosmia—loss of smell—as a specific symptom that suggested infection.
AI has the potential to truly personalize a patient’s care. We have not even scratched the surface of how AI can leverage vast amounts of data to create a clear and vibrant data picture for every patient. By combining insights from health data including clinical notes, genomic data, imaging data, and physiologic data, we can enable clinicians to make better decisions about each patient’s care more accurately and efficiently.
Multiple callers expressed well-founded concerns about the use of AI in healthcare. Some fear that AI will destroy their personal connection with their provider. I believe that connection is sacrosanct. That therapeutic doctor-patient relationship is the reason that so many of us got into healthcare in the first place. As I see it, the role of AI is to augment that relationship—not detract from it. Another caller shared the concerning fact that general AI models like ChatGPT also “hallucinate” to invent nonexistent sources and information. No physicians or scientists I know are using GPT models in any kind of informational capacity. That said, there are already companies effectively addressing the hallucination problem, and in the next five years we will likely have models that reliably deliver factually accurate information.
Another risk of AI in healthcare is encoded bias. In pregnancy care, we think about bias a lot, as the brunt of the maternal health crisis has fallen largely on the shoulders of Black and Native women, who have around 3x worse outcomes than white women in the U.S. At Delfina, when we build our models to serve these populations, we mitigate bias of the data inputs—including selection bias, as in whose data we are including, and any racial bias in how that data is collected or presented.
With AI, we have a compelling opportunity to flip the script from reactive medicine to preventative care. Especially in prenatal care, OBs are under financial pressure to see more and more patients a day, spending less time with each. Though well-intentioned, previous attempts at value-based care have fallen flat due to maternal care deserts, and physicians lacking the tools to efficiently offer each patient the whole-person care that they need. With the kinds of AI-driven tools we’re developing at Delfina, we believe that we can help reorient the healthcare system to truly adopt an outcomes-driven approach, one without the financial imperatives to grow volume or squeeze margins.
Though the use of AI in maternal health, and in healthcare in general, is still in the early stages, I believe that these tools will be monumental in transforming healthcare for the better. I am excited to see physicians, academics, entrepreneurs, and the government come together to leverage these new technologies for the benefit of patients.
I am grateful to Dr. Tignanelli, Chris Farrell, and all our callers on the program for engaging in this discussion with me. There is so much more to this timely topic, so if you are curious about AI in healthcare, listen to our full conversation here. If you want to learn more about how we are leveraging AI technology at Delfina to deliver customized, closed-loop care to pregnant patients, join us.
Last week, I spoke with Chris Farrell of MPR News...
